[AH] Red noize

[Martin Czech]

:::Im not sure I understood...
:::
:::1) The all zero's state is disallowed... I have an RC filter that will throw 
a "1" in the register when that occurs... and it DOES (every three hours or 
so...) I have never heard it... but it MUST !!!

Such shift registers get stuck at the all 0 or all 1 state (XOR XNOR).
A stable design should not get trapped, but you have to ensure initializing.


:::2) I don't get the BBD idea... you mean so you can change the clock rate??? 
you can do that with the shift register also... This is part of the Blacet "Dark 
Star" chaos generator... the SN78477 does it this way.

The idea is: short sequences sound like harmonic waves.
             very long sequneces like white noise
             
The difference is the delay (those forms with only two taps at the end).
Of course, just making the delay longer will not allwyas yield
the optimum length sequence, but this not required here anyway.

One could switch in delay stages, but may be better modulate an
analog delay line. Of course this will give problems with multirate
sampling. Anyway, the delay line carries only digital information.

If a person has already a BBD delay, an experiment would be not so difficult...


m.c.